HB100 radar!

[jgu1]

Hi Martin!

Sorry, first seen it now. Thank´s Please have a look I have just answered.

Jorgen

[jgu1]

Hi MArtin!

Please have a look in PM.

[medelec35]

Thanks Jorgen.
How are you getting on with testing?

[jgu1]

Unfortunately not enough. As I mentioned I will have made a kind of horn for the HB 100 so it will be more directional. But definitely I should have played some more with it. I'm very excited to hear your result too. when I have news you hear

Jorgen

[jgu1]

Hi Martin!

I have done some more test. As I wrote the last test with your program I saw the ligth, I could see some correct speed in display, with the first amplifier, but I have to stay exactly in front of the opject i want to measure the speed of, not so good if it is cars . But this amplifier is not stable, sometime lot of numbers in display. So I have now build the amplifier from the project with the Proff PCB you know. This is much more stable and it is possible to adjust the gain and it is possible to measure the speed from long distance.
To be sure everything work, I test it with his IDE program, I may say perfect.
Next step I will implement this amplifier (PCB one) with your FC program and test it. As soon I get five minute more free, I perform a test and you will be the first who get the result I have the feeling it will be perfect and I'm sure I'll get more accuracy with the readout in your program .

Br Jorgen.

[jgu1]

Hi Martin!

As promised, test with your FC HB100 program connected to the"PCB" Amplifier.

It work, and when you will give the HB100 a test I recommend this Amplifier. It can detect vehicles up to 100m away.

After adjusting the gain and find the rigth angle for the HB100 and good distance between the vehicle I get a nice result. I can measue the speed when the vehicles approach and drive away
I don´t know how you calculate the result, but sometime I get a strange result. Imagine a car is approach with 50km/h the first readout is maybe 130km/h then 50 km/h and then maybe 28/h It is the same with the bobblesort program. Is it possible to filter out the high and low? If not I can live with that, anyway your program also work.
Begin testing Martin.

Jorgen

[medelec35]

Thanks, Jorgen.
Have you compared it with the Arduino version?

Imagine a car is approach with 50km/h the first readout is maybe 130km/h then 50 km/h and then maybe 28/h

What you can try is to up the value of

Code: Select all

SimularValue > 20

decision branch
Try 50.
The higher the number the more the odd frequencies are rejected.
If it's taking too long then reduce the delay a bit more, e.g to 2 ms

[jgu1]

Hi Martin!

Yes I have compared it. So far I feel they are equal. When I get it 100% stable is it then possible to store maybe 10-20 measurement. Then I will drive to a place with almost no traffic and then drive by my self and take a GPS with me and then compare speed. It is a difficult when I am alone I will try to follow your advice tomorrow. It´s dark now . Now I am figthing with some display which not working.


You hear tomorrow

Jorgen

[jgu1]

Hi Martin!

Yes it help´s more stable now. Again, It is nessesary to play with the gain of the amplifier, the angle and SimularValue. I continue later, cold uotside

You hear..

Jorgen

[medelec35]

Yes it help´s more stable now.

Sounds like it's going in the right direction!
If the number within the decision branch or the delay is too high, then the event will be missed altogether.
Of course, it was only my theory, so I don't know what the highest optimum value is?
when you have found out if you could let us know, please that will be great for anyone else following this topic.